I Built a ChatGPT Clone (Here's How It Went)

This is from Week 2 of TAI's 12-week AI Engineering cohort. We asked Lovable, an AI prompt-to-app builder , to clone ChatGPT. Our instructor Dr Akshika (Aki) Wijesundara led the session, and 15 minutes later the clone was live on the internet, talking to GPT-4. Then we did it the proper way too: Next.js from scratch, raw OpenAI calls, no shortcuts. Two paths to the same app, side by side. 🎓 JOIN THE NEXT COHORT → https://theaiinternship.com/ • 12 weeks, live mentor-led sessions • Build real projects: RAG, agents, fine-tuning, MCP, capstone • Small batches, direct access to instructors • Career support + portfolio reviews ━━━━━━━━━━━━━━━━━━━━━ In this build you'll see: ✓ Lovable: prompt → working ChatGPT clone in 15 min ✓ Wiring OpenAI's chat.completions endpoint behind a real UI ✓ Connecting Supabase as the back-end ✓ Pushing to GitHub + deploying live on Vercel ✓ Same project rebuilt from scratch with Next.js + Cursor ✓ Bonus: the Streamlit version (pure Python UI) 🕐 CHAPTERS 0:00 Intro — building ChatGPT two ways 1:30 The fork: no-code vs from scratch 2:30 What is Lovable 4:00 Sending the build prompt 6:30 What is Supabase (back-end as a service) 8:00 Adding the OpenAI API key 9:30 First conversation with our clone 10:30 Pushing the project to GitHub 12:00 The Vercel deploy pipeline 14:00 The live deployed app 15:00 Building the same thing from scratch in Next.js 17:30 Letting Cursor agent generate the chat UI 20:30 Wiring the OpenAI endpoint manually 24:00 Bonus: Streamlit (Python-first UI) 27:00 How junior engineers stand out in a vibe-coding world 🛠️ TOOLS MENTIONED • Lovable — https://lovable.dev • Vercel — https://vercel.com • OpenAI Platform — https://platform.openai.com • Cursor — https://cursor.com • Supabase — https://supabase.com • Streamlit — https://streamlit.io ━━━━━━━━━━━━━━━━━━━━━ 📚 THE FULL CURRICULUM W01 — Environment Setup & Your First OpenAI Call W03 — REST APIs & JWT Auth with FastAPI W04 — Vector Databases & Embeddings W05 — Fine-Tuning Models (SFT, DPO, RFT) W07 — RAG Fundamentals W08 — LangChain & LangGraph W09 — Build an MCP Server W11–W12 — Capstone project + showcase → Apply: https://theaiinternship.com/ About TAI — The AI Internship We train engineers to ship real AI products. 12-week mentor-led cohorts, real codebases, real deployment, real career outcomes. #AIEngineering #ChatGPTClone #Lovable #NextJS #Vercel #OpenAI #VibeCoding